DESCRIPTION: Industrial noise environments typically consist of non-Gaussian, non-stationary, time varying noise. An additional complication for the assessment of hearing hazard is posed by the intermittent nature of the worker's daily exposure. There is virtually no data available in the literature on this most general and realistic type of industrial noise exposure. We propose to model the essential features of such exposure conditions and determine the audiometric and histological consequences of exposure. Data from the proposed experiments will address issues germane to our understanding of the following problematic areas in noise effects research. These are contemporary issues that are critical to the development of a comprehensive damage-risk criterion. (1) The equal energy hypothesis (EEH) which forms the basis of current damage-risk criteria for noise exposure is insensitive to the temporal features of an exposure. Under what exposure conditions is the EEH applicable? What are its limitations? Is the kurtosis metric a necessary adjunct to the Leq? (2) To what extent is noise-induced hearing loss affected by time varying and interrupted non-Gaussian exposure conditions? (3) What is the influence, if any, of the 'toughening' phenomenon, produced by interrupted exposures? Does interruption reduce noise-induced hearing loss? The chinchilla (n-150) will be used as the animal model, evoked response audiometry will be used to estimate pure tone thresholds (audiograms) and surface preparation histology will quantify cochlear sensory cell loss. The proposed experiments are an extension of our current work to more generalized and realistic exposure conditions. The data that will be acquired are expected to contribute to the development of the kurtosis metric and its application to risk assessment. All exposures will model an ideal work-day/week (8h/day with interruption, 5day/week) and last for a period of three weeks. Each exposure (either Gaussian or non-Gaussian) will be paired with an unvarying, uninterrupted control exposure (24h/day, 5d) having equivalent energy and spectrum. It is important to understand what features of realistic industrial noise environments pose the greatest hazard to hearing and to develop a reliable metric for risk assessment. A database consisting of results from approximately 150 subjects run through 18 different complex noise exposure conditions will be constructed. The statistical metric, kurtosis, the peak SPL and peak/interval histograms, the crest factor, and the pattern of temporal variation/interruption of the complex noise exposure will be varied. A large sample size and a wide variation of exposure parameters are necessary to insure statistical power for the correlations that will be made among the independent (noise parameters) and dependent (measures of trauma) variables.